If you are reading this post, then you probably have heard about “4G”, the 4th generation cellular network. For a cell phone user, 4G means improved data speeds that allow faster delivery of multimedia-based applications, see our previous post, What is 4G?, for more details. On the other hand, the network operator desires to spend a minimum on upgrading network infrastructure and prefers to buy a backhaul solution that supports current and near future capacity demands of a cellular network.
Thus, it is important to improve the capacity of wireless backhaul links. To increase transmission capacity, wider channel spacing can be used. However the wireless spectrum is expensive and may not be available in some countries. Using transmission in high frequency bands, such as 60 GHz and above, provides the bandwidth needed to increase capacity. However, very high radio frequencies increase the cost of radio components. In addition, 60 GHz links limit transmission range due to high absorption of radio waves by the atmosphere, making this solution somewhat cost inefficient. One efficient way of improving the capacity of a communication link is to increase the order of the digital communication modulation scheme used for transmission.
In simple terms, digital modulation is the process of mapping a group of data bits into an information symbol that gets transmitted, after up-conversion to the radio frequency (RF) of the link. The most popular digital modulation scheme used in wireless radios is known as quadrature amplitude modulation (QAM). For a given symbol rate, increasing the modulation order, or equivalently packing more bits per symbol, would be an effective way to increase the capacity of a microwave link. For example, each symbol in a 64-QAM signal represents 6 data bits, while for 256-QAM and 1024-QAM signals it represents 8 and 10 data bits, respectively. Therefore, 1024-QAM provides (theoretically) a 25 percent increase in capacity over 256-QAM and an impressive 67 percent increase in capacity compared to 64-QAM.
The price paid for achieving such an increase in capacity is more complex signal processing algorithms and stricter requirements for channel quality, e.g. higher signal-to-noise ratio (SNR) at the receiver is required. In that case, increasing the modulation order for some networks under normal operating conditions can have a diminishing return on throughput. This is due to the fact that the required SNR for an acceptable receiver performance rarely can be met.
Why this is the case? Let us briefly discuss the challenges in increasing the modulation order. Higher modulation order results in larger pool of symbols available for transmission. For example, for 64-QAM, there exists 64 symbols in a 2D grid (known as constellation points) compared to 1,024 symbols for 1024-QAM for the same grid size. Clearly, increasing the number of symbols (assuming fixed power) makes the symbols closer to each other in this 2D grid. Thus, data detection at the receiver becomes more susceptible to errors due to impairment.
In practical terms, receiver circuits are affected by thermal noise, clipping and non-linearity of power amplifiers, phase noise and many other distortions that are beyond the scope of this post. It is worth mentioning that increasing the signal power beyond some limits results in actually decreasing the received SNR since many of these distortions associated with RF circuits are dependent on the transmitted power. Rather, the way to increase the modulation order is to improve the detection schemes and build circuits that are less susceptible to power-related distortions, along with improving the correction mechanisms at the receiver for phase noise and other impairments.
At Aviat Networks, we have the expertise and knowledge to build the highest quality microwave radios that can work at cutting edge signaling schemes. We will make sure that our customers see a sizable return—not a diminishing one from increasing the modulation order. Our pledge is that microwave backhaul will always exceed the capacity requirements of our customers.
Senior Signal Processing Engineer, Aviat Networks
By now, you have seen the blogs, read the tweets and perhaps watched a YouTube video about “4G” mobile networks. In these postings, various claims and counterclaims have been made for what really defines 4G wireless. Further down in the industry dialogue, debate has been swirling among the ITU, IEEE 802 and various telecom analysts and pundits about what constitutes 4G. The technical acronyms LTE, WiMAX, HSPA+ and perhaps others have floated through the ether, creating more confusion than clarity.
All this happened when ITU let the genie out of the bottle in late 2010 and loosened the technical definition of what is truly 4G. The answer had been mobile technology capable of 100 Mbps+ downloads. However, ITU seems to have given mobile operators and others with vested interests enough leeway to define 4G as any mobile broadband technology that is faster than “3G,” which enjoyed a similar hype and uncertainty when it debuted in the early 2000s. And so began the public’s conditioning to equate more Gs with faster throughput.
Of course, all these Gs only refer to the generation of mobile technology, currently in its third generation in most places, with some limited availability of fourth generation technology. For the record, 4G technology in ITU’s strictest sense only refers to Long Term Evolution (LTE) Advanced and WiMAX 802.16m. Even current LTE and WiMAX 16e installations do not qualify. They are evolutionary steps on the road to 4G. And though HSPA+ is a fast download technology, it is still a third generation mobile telecom technology. Still, some HSPA+ carriers are achieving 21 Mbps downloads—faster than the 12 Mbps of early LTE carriers. With a software upgrade by the end of 2011, HSPA+ carriers can conceivably get up to 42 Mbps—but that is the theoretical maximum. Someday, LTE operators could hypothetically top out at 300 Mbps, but that day is not in the immediate future.
What is immediately apparent and most important is what 4G means to the end user. Most people cannot be bothered to dive into the technical details of mobile broadband technology, even if they are capable of grasping its intricacies. What they can grasp is faster mobile video loads with a minimum of latency and lack of jitter. What they can get is the mobile Internet displaying web pages with images in place and not red Xs or empty pictureholders. What is important is delivering content to the end user—wherever she is—faster than she expects, however many Gs it takes….
The introduction of new high-speed 4G mobile technologies is gathering pace, but there are indications that many operators still do not have a clear understanding of what impact this will have on their backhaul networks. With the forecasted demise of copper-based backhaul in U.S. networks, operators are faced with the critical decision between deploying either fiber or microwave to the cell-site.
The problem lies in the uncertainty surrounding what the true future capacity needs for 4G/LTE will be. Misreading this requirement could ultimately lead to over-building backhaul capacity, resulting in the waste of enormous amounts of network investments; money that instead could be put into more substantive ways of increasing network capacity, such as investing in new spectrum.
Driven by this uncertainty, operators risk making commitments that lock in the high lifecycle costs of building or leasing fiber – costs that far exceed that of a typical microwave connection. In reality, deploying fiber to the cellsite due to concerns about running out of backhaul capacity is a significant overkill. Based upon the 4G technologies and deployment scenarios, it is possible to predict what the maximum backhaul needs for 4G sites will be, meaning that in many cases operators can avoid a budget burden that they must live with for years to come.
So when it comes to backhaul need for 4G what is needed is a “dose of reality,” that then enables optimal backhaul network planning that balances realistic capacity expectations with total cost.
Check back for the second part of this post next week, when I take a look at some of the things that are driving this uncertainty.
Director of Marketing, Aviat Networks